Interestingly, their own pics clearly show two types of output transistors. The one set looks very convincingly like NOS B337, while the other looks like...something else. The cost of replacing the outputs is not prohibitive if they turn out to be fake.
Otoh, seeing there is no thermal compensation or effective cooling, even the modest 2W of dissipation per device may get Ge into trouble, especially in a case with poor ventilation. It will be ironic if only silicon devices survive
Bought and received the Choke Loaded (headphone) Output Amplifier (111 euro)
Made some measurements with Arta (in the attached PDF)
Build quality of this headphone - preamp is very good
Thermal stability ok. Temperature approximately 15 Celsius above room temp after 2 hours
Sound? I only have two entry level headphones (Koss Porta Pro and Sennheiser PXC 250). Sound is pleasant and non fatiguing
Limitations
As any product it has its limitations.
Output both in terms of available voltage swing and power is not phenomenal. Thus it will not drive Planar headphones or a Mosfet Follower power amp to satisfactory level.
The company mentions that the amp may come with different Output transistors. Results therefore may vary. But i have the impression that they are really dedicated to deliver a good product
I decided to jump on this germanium bandwagon because being that I really enjoy working with tubes it's a bit hypocritical to make fun of obsolete semiconductors. That said, there was a link posted to an eBay auction for a lot of house numbered Bendix transistors. It looks like the seller split the lot in half and it automatically made me an offer so here we are.
A quick mock up for measuring hFE and leakage using this setup http://www.geofex.com/Article_Folders/ffselect.htm gives the following results. Half are duds with no gain, a few are dead, and the rest vary wildly.
https://docs.google.com/spreadsheet...RL6zYOxgghsTrwZBzTM8ltRf7X6pIatEtwjmz/pubhtml
A friend of mine has a Tek curve tracer so expect some more data.
A quick mock up for measuring hFE and leakage using this setup http://www.geofex.com/Article_Folders/ffselect.htm gives the following results. Half are duds with no gain, a few are dead, and the rest vary wildly.
https://docs.google.com/spreadsheet...RL6zYOxgghsTrwZBzTM8ltRf7X6pIatEtwjmz/pubhtml
A friend of mine has a Tek curve tracer so expect some more data.
With the new forum I do not find a way to quote from two messages.
See below.
Ok finally got a chance to play with the curve tracers. First I tried the Tek 571 which seemed a bit noisy or perhaps was just having a bad day. Then we busted out the tube based Tek 575 and the plots also look pretty poor. For reference a 2N3906 looked flat and linear as ever on the 575. My buddy thinks these transistors are either rejects or designed for switching purposes. The lot seemed to have different date codes and slight differences on package stamping. We decided to cut open one of the bad ones and you can see the massive disc of germanium inside.
Oh and on the Tek 575 you can hold the transistor in your hand and watch the plots drift right off the screen!
Oh and on the Tek 575 you can hold the transistor in your hand and watch the plots drift right off the screen!
Attachments
Could you show Bendix Ge photo with Tek 575 ?
Maybe is time now to find New Old Stock audio grade germanium, small signal and power transistors ?
Test them with correct mica and goop and heatsink included
Have a nice day
Could you show Bendix Ge photo with Tek 575 ?
Maybe is time now to find New Old Stock audio grade germanium, small signal and power transistors ?
Test them with correct mica and goop and heatsink included
Those photos were all different samples of the Bendix transistors. Here is one more showing the 575 controls. I did have the current cranked up for one but the plots didn't look any better. Maybe next time I'll see how some soviet transistors look. MP10, MP26, 1T806B, P206.
I don't need any more test equipment but a curve tracer is calling my name...
Attachments
Just wanted to say I prototyped this circuit (minus the 4700uf rail caps) and on my linear bench supply the performance is pretty poor. The output transistors don't have heatsinks so I can only test for 30 seconds or so before they get hot.
At 1v input it's putting out 6.5 watts into 8 ohms at 5.0% THD. The DC offset at the load is 0.7v. Once the heatsinks arrive I can do some more tweaking and see what's going on. Amazon sells heatsinks for solid state relays and they seem like a good fit.
Attachments
You can find out more here.
https://ldsound.ru/jlh-germanium-transistor-amp/
I didn't make this circuit. The amplifier works in class "A", so you need to use the appropriate heatsinks.
You didn't even get as far as my second paragraph. I know the transistors need heatsinks but don't have them yet. With the link jpatay provided I now know that this amplifier is only rated for 5-6 watts and it was into clipping. Reducing the input to 0.7v reduces the distortion to 0.2%. Once the heatsinks arrive I will do a better analysis.